17 Commits

Author SHA1 Message Date
Eugene Zelenko
cecd8f18e2 [Analysis, Transforms] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@312383 91177308-0d34-0410-b5e6-96231b3b80d8
2017-09-01 21:37:29 +00:00
Alexey Bataev
a64c8c9af1 [SLP] General improvements of SLP vectorization process.
Patch tries to improve two-pass vectorization analysis, existing in SLP vectorizer. What it does:

1. Defines key nodes, that are the vectorization roots. Previously vectorization started if StoreInst or ReturnInst is found. For now, the vectorization started for all Instructions with no users and void types (Terminators, StoreInst) + CallInsts.
2. CmpInsts, InsertElementInsts and InsertValueInsts are stored in the
array. This array is processed only after the vectorization of the
first-after-these instructions key node is finished. Vectorization goes
in reverse order to try to vectorize as much code as possible.

Reviewers: mzolotukhin, Ayal, mkuper, gilr, hfinkel, RKSimon

Subscribers: ashahid, anemet, RKSimon, mssimpso, llvm-commits

Differential Revision: https://reviews.llvm.org/D29826

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@310260 91177308-0d34-0410-b5e6-96231b3b80d8
2017-08-07 15:25:49 +00:00
Alexey Bataev
20e83eb193 Revert "[SLP] General improvements of SLP vectorization process."
This reverts commit r310255.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@310257 91177308-0d34-0410-b5e6-96231b3b80d8
2017-08-07 14:51:52 +00:00
Alexey Bataev
4362efa895 [SLP] General improvements of SLP vectorization process.
Summary:
Patch tries to improve two-pass vectorization analysis, existing in SLP vectorizer. What it does:
1. Defines key nodes, that are the vectorization roots. Previously vectorization started if StoreInst or ReturnInst is found. For now, the vectorization started for all Instructions with no users and void types (Terminators, StoreInst) + CallInsts.
2. CmpInsts, InsertElementInsts and InsertValueInsts are stored in the array. This array is processed only after the vectorization of the first-after-these instructions key node is finished. Vectorization goes in reverse order to try to vectorize as much code as possible.

Reviewers: mzolotukhin, Ayal, mkuper, gilr, hfinkel, RKSimon

Subscribers: ashahid, anemet, RKSimon, mssimpso, llvm-commits

Differential Revision: https://reviews.llvm.org/D29826

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@310255 91177308-0d34-0410-b5e6-96231b3b80d8
2017-08-07 14:03:17 +00:00
Alexey Bataev
340067cec4 [SLPVectorizer] Generalize interface of functions, NFC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@309816 91177308-0d34-0410-b5e6-96231b3b80d8
2017-08-02 14:38:07 +00:00
Alexey Bataev
6ed2765ff1 [SLP] Comment fix, NFC.
Fixed comment in function description.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304940 91177308-0d34-0410-b5e6-96231b3b80d8
2017-06-07 20:37:24 +00:00
Adam Nemet
65ad27f81e [SLP] Emit optimization remarks
The approach I followed was to emit the remark after getTreeCost concludes
that SLP is profitable.  I initially tried emitting them after the
vectorizeRootInstruction calls in vectorizeChainsInBlock but I vaguely
remember missing a few cases for example in HorizontalReduction::tryToReduce.

ORE is placed in BoUpSLP so that it's available from everywhere (notably
HorizontalReduction::tryToReduce).

We use the first instruction in the root bundle as the locator for the remark.
In order to get a sense how far the tree is spanning I've include the size of
the tree in the remark.  This is not perfect of course but it gives you at
least a rough idea about the tree.  Then you can follow up with -view-slp-tree
to really see the actual tree.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@302811 91177308-0d34-0410-b5e6-96231b3b80d8
2017-05-11 17:06:17 +00:00
Sanjoy Das
399b4d037d Rename WeakVH to WeakTrackingVH; NFC
This relands r301424.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@301812 91177308-0d34-0410-b5e6-96231b3b80d8
2017-05-01 17:07:49 +00:00
Sanjoy Das
263da12ab2 Reverts commit r301424, r301425 and r301426
Commits were:

"Use WeakVH instead of WeakTrackingVH in AliasSetTracker's UnkownInsts"
"Add a new WeakVH value handle; NFC"
"Rename WeakVH to WeakTrackingVH; NFC"

The changes assumed pointers are 8 byte aligned on all architectures.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@301429 91177308-0d34-0410-b5e6-96231b3b80d8
2017-04-26 16:37:05 +00:00
Sanjoy Das
d0cf26e443 Rename WeakVH to WeakTrackingVH; NFC
Summary:
I plan to use WeakVH to mean "nulls itself out on deletion, but does
not track RAUW" in a subsequent commit.

Reviewers: dblaikie, davide

Reviewed By: davide

Subscribers: arsenm, mehdi_amini, mcrosier, mzolotukhin, jfb, llvm-commits, nhaehnle

Differential Revision: https://reviews.llvm.org/D32266

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@301424 91177308-0d34-0410-b5e6-96231b3b80d8
2017-04-26 16:20:52 +00:00
Daniel Jasper
8de3a54f07 Revert @llvm.assume with operator bundles (r289755-r289757)
This creates non-linear behavior in the inliner (see more details in
r289755's commit thread).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@290086 91177308-0d34-0410-b5e6-96231b3b80d8
2016-12-19 08:22:17 +00:00
Hal Finkel
bffeba468d Remove the AssumptionCache
After r289755, the AssumptionCache is no longer needed. Variables affected by
assumptions are now found by using the new operand-bundle-based scheme. This
new scheme is more computationally efficient, and also we need much less
code...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@289756 91177308-0d34-0410-b5e6-96231b3b80d8
2016-12-15 03:02:15 +00:00
Alexey Bataev
62c10d6781 [SLPVectorizer] Improved support of partial tree vectorization.
Currently SLP vectorizer tries to vectorize a binary operation and dies
immediately after unsuccessful the first unsuccessfull attempt. Patch
tries to improve the situation, trying to vectorize all binary
operations of all children nodes in the binop tree.

Differential Revision: https://reviews.llvm.org/D25517

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@288115 91177308-0d34-0410-b5e6-96231b3b80d8
2016-11-29 08:21:14 +00:00
Chad Rosier
454a60a86c [SLP] Don't pass a global CL option as an argument. NFC.
Differential Revision: https://reviews.llvm.org/D24199

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@280527 91177308-0d34-0410-b5e6-96231b3b80d8
2016-09-02 19:09:50 +00:00
Chad Rosier
5f960b8b75 [SLP] Arguments should be camel case, and start with an upper case letter. NFC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@280248 91177308-0d34-0410-b5e6-96231b3b80d8
2016-08-31 15:06:58 +00:00
Sean Silva
de764bb96b Fix up an include guard.
This should have been done as part of the move in r274960.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274999 91177308-0d34-0410-b5e6-96231b3b80d8
2016-07-09 22:56:39 +00:00
Sean Silva
eb2d4257f4 [PM] Fix a think-o. mv {Scalar,Vectorize}/SLPVectorize.h
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274960 91177308-0d34-0410-b5e6-96231b3b80d8
2016-07-09 03:11:29 +00:00